Hello, to detect gene protein entities you need to train your own model. Have a look at BioNLP 2004, as far as I know this is Genia corpus data, OpenNLP already includes format support for it and you should be able to train on it.
You might need to adjust the feature generation to make it perform well. HTH, J?0?2rn On 07/09/2013 09:41 AM, ???????? wrote:
Dear Sir, I am Tong WEI, a Master student from Lanzhou University, China. I am writing to you concerning the use of OpenNLP. I got a problem when I use the command line tool to try out the Name Finder since I'm new to OpenNLP tool. In the model download page, English Name Finder models just include date name finder model, location name finder model, money name finder model, organization name finder model, person name finder model and time name finder model. In my project, I want to recognize the gene and protein named entities in the text, however, the Name Finder can't detect the gene or protein entities in my text. So my question is that is there gene name finder model or protein name finder model in the English Name Finder models? Thank you so much for your time and best regards. Miss Tong WEI, Master Candidate School of Information Science & Technology Lanzhou University, P.R. China
